Genes which control early steps during mammalian CNS development have not yet been described. There is strong, but indirect evidence that homeobox genes may have this property. These genes share a region, the homeobox, which is found in many of the development regulating genes of the fruitfly Drosophila melanogaster. This region encodes a sequence specific DNA binding domain. The insect genes are known to encode transcription factors which control expression in the CNS and elsewhere, of other genes in a body position sensitive manner. The expression of the murine Hox genes in the developing CNS also suggests that these genes are also sensitive to positional information. Initially, the characterization of the Hox 1.2 gene and its product will be completed, including testing whether it is also able to bind to DNA. To determine whether the murine Hox 1.2 and Hox 1.3 proteins are transcription factors, several potential regulatory targets will be screened for responsiveness to these proteins: 1) as predicted on the basis of its DNA binding specificity, the Hox 1.3 protein appears to affect expression of herpes simplex immediate early genes. This will be tested directly at the molecular level to determine whether the herpes gene regulatory elements respond to Hox proteins. 2) Since Drosophila homeodomain proteins regulate their own and each other's expression, likely endogenous targets of Hox proteins include the Hox genes themselves. This will first entail completing the characterization of the Hox 1.2 and Hox 1.3 transcription units to define their regulatory domains. These regions will then be tested to determine whether they respond to Hox proteins. After the likely confirmation that Hox proteins are transcription factors, it will then be more meaningful to begin studies to establish whether Hox 1.2 or Hox 1.3 are important during CNS development. Hox 1.3 expression in Purkinje cells will be characterized throughout development of normal mice and of mutant mice with cerebellar defects. These observations will be extended to analysis of neonatal cerebellar explants where morphogenesis can be studied under controlled conditions. Homeobox gene expression is responsive to retinoic acid and certain peptide growth factors. The effects of these morphogens and teratogens upon cerebellar morphogenesis will be studied using these explants. Any morphological changes will be correlated with changes in homeobox gene expression. These studies should give insight into mechanisms important during both normal and disturbed embryogenesis.